Balancing Sustainability and Profitability: The Financial Effect of Green Innovation in Chinese High-Pollution Industries

Abstract

Green innovation plays a crucial role in sustainable development, yet its financial impact on high-pollution industries remains underexplored. This study analyzes the short- and long-term financial effects of green innovation using 30,108 firm-year observations from Chinese A-share listed companies in high-pollution industries (2009–2022). Employing fixed-effects regression models, green innovation is measured through environmental patents (EnvrPats) and environmentally innovative patents (EnvrInvPats), with Return on Assets (ROA) as the financial performance metric. To address endogeneity concerns, instrumental variable (IV) techniques are applied using digital transformation (DT) as an instrument, alongside a two-stage Generalized Method of Moments (GMM) approach for validation. This study explores the moderating roles of Sustainable Liquidity Reserves (cash flow) and the Sustainable Development Index (ESG performance), while a channel test examines the influence of R&D expenditures and financial constraints. A heterogeneity analysis reveals that firms in high-pollution industries experience greater short-term financial benefits from green innovation, driven by regulatory pressures and public scrutiny. A pre- and post-COVID-19 analysis highlights the increased importance of green innovation in firm resilience during economic disruptions. Robustness checks, including alternative financial performance measures and nonlinear modeling, confirm the reliability of the findings. While green innovation imposes initial financial costs, firms with stronger cash reserves and ESG performance can better absorb these costs and achieve long-term financial gains, emphasizing the need for targeted policy support to facilitate sustainable growth.

1. Introduction

In recent years, sustainable development has become a critical focus for businesses and policymakers, as environmental degradation and resource depletion pose significant threats to global well-being [1,2]. The increasing global pressure for environmental protection, driven by rapid economic and population growth, has led to a surge in green innovation as a key strategy for firms aiming to balance profitability with environmental responsibility [3,4]. One of the most efficient methods for firms to tackle climate change issues is through the adoption of green innovation, which entails the creation of eco-friendly products, services, and processes that reduce the adverse effects of business activities on the environment. Green innovation has garnered significant attention in recent years as firms worldwide endeavor to reconcile economic growth with environmental sustainability. Although numerous studies have investigated the capacity of green innovation to augment company performance via enhanced operational efficiency and regulatory adherence, a significant gap persists in comprehending the short-term financial effects [5,6], especially within high-pollution sectors. Despite its growing relevance, the short-term financial impact of green innovation remains underexplored, particularly in industries facing stringent environmental regulations and operational constraints [7,8]. Moreover, this study aims to address these gaps by exploring the dynamic financial relationship between green innovation and firm performance, with a focus on the role of financial flexibility and industry-specific factors. Nonetheless, whereas the long-term advantages of green innovation are evident, including enhanced market positioning, diminished waste, and adherence to environmental regulations, the short-term financial implications are more difficult.

Firm financial performance, commonly measured by indicators such as Return on Assets (ROA), reflects a company’s efficiency in utilizing its assets to generate earnings and is widely used in sustainability and innovation-related empirical research [7,8]. In the context of green innovation, financial performance serves as a critical benchmark for understanding whether environmentally sustainable practices align with or hinder short-term profitability [9]. While green innovation often improves regulatory compliance and operational efficiency in the long run, its immediate impact on financial performance remains ambiguous, particularly in high-pollution industries where compliance costs and R&D investment are substantial [10,11]. Firms investing in eco-innovations may experience temporary declines in ROA due to upfront costs, longer innovation cycles, and uncertain returns, making it essential to examine both the short- and long-term financial implications [9]. As such, this study positions ROA as a key indicator to evaluate how green innovation initiatives affect corporate profitability in China’s heavily polluting sectors, contributing to a more nuanced understanding of the sustainability profitability nexus.

Green innovation has increasingly been recognized as a vital corporate strategy for aligning environmental responsibility with firm performance, particularly in pollution-intensive industries [12,13]. Prior studies emphasize the long-term benefits of green innovation such as improved regulatory compliance, cost efficiency, and enhanced firm reputation [14]. However, the financial impact of such innovation in the short term remains contested, especially in contexts where green initiatives require substantial investment in R&D and compliance-related upgrades [15]. The research gap lies in the limited understanding of how green innovation affects short-term financial performance, particularly in developing countries and high-pollution sectors where cost pressures are more severe [16]. The objective of this study is to advance empirical insight into the financial implications of green innovation by evaluating its short-term effects on firm profitability thereby contributing to ongoing debates on whether sustainable practices align with or conflict with near-term economic performance.

While existing literature has explored the direct effects of green innovation on firm outcomes, limited attention has been given to how internal firm characteristics shape this relationship, especially in high-pollution industries of developing countries [17]. In particular, financial flexibility, reflected through cash flow (Sustainable Liquidity Reserves), may influence a firm’s capacity to manage the short-term costs associated with green innovation [18]. Similarly, sustainability performance, measured through ESG indicators, reflects stakeholder alignment and environmental commitment, but may also impose additional financial burdens in the short run [19]. Despite their growing relevance, empirical studies examining the moderating roles of these two factors in the green innovation–financial performance relationship remain scarce [15]. Therefore, the second objective of this study is to investigate how financial flexibility (cash flow) and ESG performance moderate the relationship between green innovation and firm financial performance in high-pollution industries. This aims to fill a key gap in the literature by revealing how firm-level characteristics either amplify or mitigate the financial impact of green initiatives, offering important insights for managers seeking to align sustainability with profitability.

Finding a bridging channel between green innovation and firm financial performance is essential to understanding how sustainability strategies translate into measurable economic outcomes [20,21]. While green innovation has been linked to various long-term advantages, limited research has explored the internal mechanisms through which it impacts short-term financial performance, particularly in the context of developing countries and high-pollution industries [10,22]. Among the potential channels, R&D expenditure often imposes financial strain in the short term due to high investment and delayed returns, whereas financial constraints can limit a firm’s ability to sustain innovation, weakening its intended financial benefits [15]. The research gap here concerns the lack of studies investigating the mediating mechanisms specifically R&D expenditure and financial constraints through which green innovation impacts firm financial performance in capital-intensive and regulation-sensitive industries [16]. The third objective of this study is to explore these mediating effects to better understand how internal financial dynamics shape the sustainability–performance relationship in emerging market firms.

This study makes several contributions to the existing literature on green innovation and firm performance. First, it provides new empirical evidence on the short- and long-term financial effects of green innovation, highlighting that while green initiatives may negatively impact short-term profitability, they are associated with positive financial returns in the long run [7,8]. This finding aligns with the Porter Hypothesis, which posits that environmental regulation can stimulate innovation that enhances firm competitiveness. Second, it demonstrates that financial flexibility (cash flow) and sustainability performance (ESG scores) play important moderating roles, where firms with stronger cash positions and higher ESG ratings are better able to manage the financial pressures of green innovation [12,13]. Firms with greater liquidity and stronger ESG profiles are better positioned to absorb the costs of green innovation and leverage them strategically—findings that are consistent with Financial Slack Theory and Stakeholder Theory, both of which emphasize the importance of internal resources and stakeholder alignment in achieving sustainable performance. Third, this study identifies R&D expenditure and financial constraints as key channels through which green innovation influences profitability, showing that while higher R&D investment may temporarily reduce financial performance, it contributes positively over time, and that tighter financial constraints hinder innovation outcomes [6]. This contributes to the Resource-Based View (RBV) by highlighting how intangible assets such as R&D and financial resilience shape competitive advantage. Finally, the results reveal heterogeneous effects, with firms in high-pollution industries benefiting more from green innovation due to regulatory incentives, and with stronger positive effects observed after the COVID-19 pandemic, when sustainability became more significant to corporate resilience strategies.

We analyze an unbalanced panel dataset of 30,108 firm-year observations from Chinese A-share listed firms in high-pollution industries between 2009 and 2022, sourced from the CNRDS and CSMAR databases. In line with prior studies that acknowledge the potential endogeneity between innovation decisions and firm performance [20], we employ an instrumental variable approach using digital transformation a factor shown to influence innovation behavior but not directly tied to short-term profitability as an exogenous instrument. This approach strengthens the causal interpretation of our findings. The results reveal that green innovation negatively affects short-term financial performance due to high initial costs, but over time, firms experience positive financial returns, especially in heavily regulated, high-pollution sectors. A U-shaped relationship is identified, indicating that while early-stage investments may reduce profitability, sustained green innovation ultimately enhances firm performance. These results underscore the strategic importance of long-term commitment to green innovation for achieving both environmental and financial goals. The findings provide practical insights for corporate managers, investors, and policymakers by illustrating how firms can navigate the trade-offs between environmental sustainability and financial performance amid increasing regulatory and societal pressure.

The sections of this study are as follows: The theoretical framework and literature review are presented in Section 2, detailing essential concepts and studies. Section 3 provides a detailed explanation of the data sample and the methods employed in the research design. Section 4 presents the empirical study, while Section 5 explores the policy implications. Section 6 encapsulates the key contributions and provides guidance for future inquiries.

2. Literature Review and Hypothesis Development

2.1. Theoretical Background

This study is based on four main theoretical frameworks. Porter’s hypothesis suggests that thoughtfully designed environmental regulations can drive innovation, resulting in increased competitiveness and financial performance [23]. According to this hypothesis, stringent environmental regulations force firms to innovate resulting in both compliance and long-term profitability [21,24]. For high-pollution industries, this theory is particularly relevant, as firms are subject to increasingly stringent regulations to reduce emissions and waste [25,26]. Implementing green innovations allows companies to meet regulatory requirements while increasing operational efficiency, reducing costs, and opening new market opportunities. This study draws on the Porter Hypothesis to explore whether green innovation, despite its short-term costs, eventually leads to long-term financial gains.

The Resource-Based View of the firm highlights that organizations achieve competitive advantage through the development and utilization of different resources and capabilities [16]. Green innovation can be a strategic asset that offers organizations a competitive edge, particularly in sectors where adherence to environmental regulations is essential [27]. Organizations focused on creating innovative eco-friendly technologies and methodologies have the potential to differentiate themselves in the market, draw in investors and consumers who prioritize sustainability, and enhance their financial results [23,28]. This study applies the RBV to argue that green innovation, as a valuable and rare capability, enables firms with high pollution to achieve both sustainability and profitability over time.

Stakeholder Theory posits that organizations should consider the interests of a diverse range of stakeholders such as shareholders, employees, customers, regulators, and the environment [29]. Organizations that demonstrate good performance in environmental, social, and governance aspects are better equipped to meet the expectations of their stakeholders who are progressively advocating for sustainable business practices [30,31]. This theory suggests that organizations with elevated ESG scores are more likely to successfully implement green innovations because they are able to align stakeholder interests while eliminating short-term financial impacts. This study uses Stakeholder Theory to explain how companies with strong ESG performance may be better equipped to meet the challenges of green innovation, particularly in highly regulated sectors.

Financial Slack Theory suggests that firms with greater financial flexibility (or slack) are better able to invest in innovation and manage short-term financial risks [19]. Cash flow as a key measure of financial slack, allows to allocate resources to green innovation without jeopardizing their operational stability [32,33]. This study uses Financial Slack Theory to explain why the moderating role of cash flow is better equipped to absorb the initial financial burdens associated with green innovation and more likely to realize long-term financial benefits.

There is a lack of research on the short-term financial challenges firms face due to green innovation investments in high-pollution industries. Additionally, the potential nonlinear financial impacts of such innovation and the role of financial flexibility have been underexplored. This study addresses these gaps through a detailed analysis of industry-specific factors.

2.2. Formulation of Hypothesis

2.2.1. Green Innovation and Firm Performance

Green innovation and growth are clearly the way of the future, and policy support for them is growing [34]. The compatibility between a wide macro-level growth strategy and micro-level incentives remains a subject of ongoing criticism. Stated differently, does green innovation assist certain companies financially or economically? A recent wave of writing has begun to investigate this subject in depth. According to [34], profitability and investment in green innovation are interconnected with sustainability. The authors discovered that societal and economic pressure to achieve sustainable growth is what propels green innovation. According to [35], organizations make decisions about green innovation based on both internal and external factors. This finding emphasizes the need of need of marker-driven processes in promoting environmentally friendly innovation. They discovered evidence in favor of the so-called Porter Hypothesis and proposed that businesses can gain indirectly from environmental performance gains in economic performance. However, they claim that just 19% of businesses that invest in green initiatives perform better, but green investment in the other 81% of businesses has negligible or even adverse effects on recurrence [36].

A resource that differs from other kinds of inventions is green innovation [37]. Environmental concerns are one of the distinctive features of this kind of innovation that drives the growth of a business’s activities along the value chain [37]. The creation and use of goods, services, procedures, and business plans that lessen their adverse effects on the environment and advance sustainable development is what we refer to as green innovation in this study. Top management is encouraged to integrate environmental innovations into the company’s commercial plan by the presence of stringent regulatory pressure and increased customer attention to innovations in a company’s ecological behavior [37,38]. Numerous studies have examined the connection between green innovation and financial performance, with varying degrees of success. Studies like the one in [39] emphasize that green innovation enhances firm performance by improving operational efficiencies and creating new market opportunities [39]. These studies align with the Porter Hypothesis, which suggests that stringent environmental regulations can stimulate innovation, leading to long-term profitability. However, other research highlights the significant initial costs associated with green innovation, particularly in research and development (R&D) and compliance measures [10,40]. These studies argue that the financial benefits of green innovation may not materialize in the short term and may even negatively impact profitability in the early stage. Empirical evidence suggests that green innovation imposes substantial initial costs, particularly in research and development (R&D), compliance measures, and technology investments.

The research indicates that companies pursuing green innovation often face transient financial difficulties, which appear in decreased profitability metrics like Return on Equity (ROE) and Return on Assets (ROA) [7,10,26]. These adverse effects are attributed to the high upfront costs of green initiatives, which do not immediately yield financial returns. Empirical studies on Chinese firms, such as those by [10,41], further confirm that the initial financial performance impact of green innovation can be harmful, particularly in heavily polluting industries where compliance costs are high. Despite the extensive research, the literature still needs to be divided on the exact nature of the relationship between green innovation and financial performance. A nonlinear or U-shaped association is shown by certain studies [42]. Other studies report a linear positive relationship [10,43]. The purpose of this study is to close this gap in understanding by presenting actual data on the short- and long-term impacts of green innovation on financial performance, with a particular emphasis on highly polluting industries. Therefore, this research suggests the following hypothesis:

H1: 

Green innovation negatively impacts a firm’s financial performance in high-pollution and low-pollution industries.

2.2.2. Mechanism Perspectives of R&D Expenditure and Financial Constraints in the Relationship Between Green Innovation and Firm Performance

Green innovation investing in research and development (R&D) is essential for green innovation, as it enables firms to develop new technologies, processes, and products that reduce environmental impact. However, these R&D investments are often associated with high upfront costs that can temporarily strain a firm’s financial performance before yielding measurable returns [11,44]. Despite these short-term financial challenges, firms that strategically invest in green R&D may experience long-term benefits, including improved efficiency, regulatory compliance, and competitive advantage in green markets [45]. This aligns with the Resource-Based View (RBV), which suggests that unique capabilities like green technologies can create sustained competitive advantages [16]. However, while the importance of R&D for innovation is well documented, its specific role in linking green innovation to financial performance, particularly in high-pollution industries, has been less explored, and the influence of financial constraints on R&D investment remains an area needing further investigation.

Given the significant role of R&D in green innovation and the influence of financial constraints, it is essential to consider these factors as channels in the relationship between green innovation and financial performance [46]. R&D expenditures can be viewed as a strategic investment, essential for achieving the long-term benefits of green innovation, but one that imposes short-term financial pressures. Firms with higher R&D spending in green innovation may experience initial financial strain, as revenues from green technologies take time to materialize [7]. However, these investments can yield significant long-term benefits, including cost savings, regulatory compliance, and improved market positioning in sustainable industries [45]. On the other hand, financial constraints, as measured by the KZ index, can exacerbate the challenges associated with green innovation [47]. Firms facing greater financial constraints may be unable to sustain their R&D investments, leading to reduced innovation capacity and diminished long-term competitiveness [11,44]. High financial constraints could force firms to forgo or delay green innovation projects, limiting the positive impact on financial performance [45]. Thus, financial constraints act as a key channel that shapes how green innovation affects firm performance, particularly in industries with high capital requirements and regulatory pressures [44]. By investigating both R&D expenditures and financial constrain as channels, this study aims to provide a more comprehensive understanding of the mechanisms through which green innovation impacts financial flexibility in overcoming the short-term challenges associated with green innovation, particularly for firms in high-pollution industries that face intense regulatory scrutiny. Based on the literature, we propose the following hypotheses:

H2a: 

The relationship between green innovation and firm performance is significantly channeled by R&D expenditure.

H2b: 

The relationship between green innovation and firm performance is significantly channeled by financial constraints.

2.2.3. The Moderating Role

The Moderating Role of Cash Flow Sustainable Liquidity Reserves

Firms with greater financial flexibility, measured through Sustainable Liquidity Reserves, are better positioned to manage the short-term financial strain associated with green innovation investments [19], also referred to as financial slack, are better positioned to invest in innovation, including green innovation, without jeopardizing their short-term financial stability [43]. Financial slack allows firms to take on the higher costs associated with research, development, and the implementation of green technologies [48]. As firms in high-pollution industries invest heavily in sustainable practices, those with substantial cash flow have the flexibility to absorb the short-term financial strain of these investments. However, there is a potential downside: firms with more financial resources may over-invest in green innovation, leading to a temporary financial decline due to the large upfront costs. Previous research highlights that while cash flow enables firms to pursue green initiatives, the financial benefits may not be immediately realized [49,50]. This study posits that firms with higher Sustainable Liquidity Reserves are better equipped to absorb the initial financial costs of green innovation, thereby reducing the adverse effects on short-term financial metrics.

H3a: 

The relationship between green innovation and financial performance is significantly moderated by Sustainable Liquidity Reserves in high-pollution and low-pollution industries.

The Moderating Role of Sustainable Development Index ESG Performance

Firms with a high Sustainable Development Index, reflecting strong ESG performance, are under greater scrutiny from stakeholders such as investors, regulators, and consumers, who demand leadership in sustainable practices [29] and expect these firms to excel in sustainable practices [33]. Firms with strong ESG scores are typically more committed to ethical and sustainable business models, which attract long-term investments and enhance reputational capital [51,52]. However, firms with higher ESG performance also face increased costs because they are often expected to go beyond simple regulatory compliance and implement more comprehensive environmental and social initiatives [32,33]. This can increase short-term financial pressures, but the long-term reputational benefits and enhanced investor trust can offset these costs, leading to better financial performance over time.

Empirical studies have demonstrated that while firms with higher ESG scores may face greater financial burdens in the short term [52], they tend to outperform their peers in the long term due to stronger stakeholder relationships and investor trust [33]. As green innovation aligns closely with ESG principles [53], firms with high ESG performance are likely to see greater scrutiny in their green innovation efforts but may achieve superior financial results as their sustainable practices enhance their market reputation [33]. While achieving high scores on the Sustainable Development Index may initially increase costs due to comprehensive sustainability initiatives, the long-term benefits include enhanced reputational capital, greater investor trust, and improved financial performance. This study examines how a high Sustainable Development Index moderates the relationship between green innovation and financial performance, suggesting that firms with strong ESG performance can leverage their sustainable practices to achieve superior financial outcomes, despite initial financial pressures. This study fills a critical gap by examining how these two factors interact and influence the financial outcomes of green innovation across industries, offering a deeper understanding of how firms can strategically manage both financial and ethical resources to achieve sustainable growth.

H3b: 

The relationship between green innovation and financial performance is significantly moderated by the Sustainable Development Index in high-pollution and low-pollution industries.

2.2.4. Nonlinear Impact

Resource Based Theory suggests that firms need to reconfigure their resources to adapt to changing environments, such as increased environmental regulations [54,55]. Empirical studies reveal that the relationship between green innovation and financial performance is nonlinear [56], with initial investments leading to negative outcomes before turning positive once firms reach a critical threshold of green technological investments [22]. This U-shaped relationship is consistent with the RBV, which emphasizes the need for sustained investments and adaptability to overcome the initial financial strain of green innovation and eventually achieve positive financial outcomes [22]. There is growing recognition in the literature that the relationship between green innovation and financial performance may be nonlinear. Some studies suggest that green innovation initially imposes financial costs, but once firms reach a certain threshold, the financial benefits begin to outweigh the costs [22]. This “U-shaped” relationship has not been fully explored, particularly in terms of identifying the tipping point where green innovation starts to have a positive impact on financial outcomes.

This research aims to address this gap by examining the nonlinear effects of green innovation on financial performance, particularly focusing on identifying the threshold where the negative impacts on profitability begin to reverse. Therefore, there is a nonlinear relationship between green innovation and financial performance, where initial investments may decrease profitability, but sustained and scaled investments eventually lead to positive financial outcomes. This hypothesis examines nonlinear effects and long-term benefits, revealing threshold levels where green innovation investments transition from having negative to positive financial impacts.

This study is supported by an integrated conceptual framework (Figure 1) drawing from the Porter Hypothesis, Stakeholder Theory, and Financial Slack Theory to examine how green innovation affects firm financial performance in high-pollution industries. The Porter Hypothesis posits that stringent environmental regulations can stimulate innovation, which may offset compliance costs and enhance competitiveness [17]. While previous research has largely emphasized the long-term strategic benefits of green innovation, fewer studies have examined its short-term financial effects, especially in developing-country contexts [25,26]. Stakeholder Theory [29] suggests that firms adopt environmentally responsible strategies to align with stakeholder expectations, and strong ESG performance may help firms manage reputational and financial risks [30,31]. Additionally, Financial Slack Theory II [19] emphasizes the role of internal financial flexibility such as available cash flow as a critical resource that enables firms to undertake costly innovation projects and buffer the associated short-term financial strain [32,33]. Additionally, the Resource-Based View (RBV) provides a foundation for understanding how firm-specific capabilities, such as R&D intensity and ESG performance, function as strategic resources that contribute to competitive advantage and performance differentiation [16]. Within this integrated framework, this study investigates not only the direct effect of green innovation on financial outcomes but also how this relationship is moderated by financial flexibility and ESG orientation and mediated through R&D expenditure and financial constraints, offering a nuanced view of the sustainability–performance nexus.

3. Data and Methodology

3.1. Data and Sample Description

This research analyzes explicitly firms listed in the Chinese A-share market. The sample period spans from 2009 to 2022, depending upon the availability of green patents and associated data. This study utilizes unbalanced panel data comprising 30,108 valid observations collected by sorting and filtering. This study’s data sources comprised two components: corporate green innovation data from CNRDS and corporate financial and other characteristics data from CSMAR. Sample data from listed enterprises in the financial sector were excluded due to their minimal contribution to green innovation initiatives [57]. This study examined listed firms identified with the symbols PT, *ST, ST, or PT, which were excluded from the research sample. In Appendix A, variables are briefly described (

Table A1. Variable Description.

Variable	Abbreviation	Measurement	Source
Financial Performance	(ROA)	Net Income/Total Assets	CSMAR
Green Innovation	GI	EnvrPats: Sum of green invention and green utility model patent applications, log-transformed EnvrInvPats: Count of green invention patent applications, log-transformed	WIND
R&D Expenditure	lnRD	Ratio of R&D Expenditure to Total Sales/Revenue	CSMAR
Financial Constraints	FC	KZ Index: Based on cash flow, market value, and leverage	CSMAR
Cash Flow (CFO)	CFO	Operating Cash Flow/Total Assets	CSMAR
ESG Performance	ESG	ESG score: Higher scores indicate better performance	CSMAR
Firm Size	Size	Natural logarithm of total assets	CSMAR
Leverage	LEV	Total Debt/Total Assets	CSMAR
Sales Growth		Annual percentage increase in sales	CSMAR
Firm Age	AGE	Number of years since establishment	CSMAR
Market-to-Book Ratio	MB	Market Value/Book Value of Equity	CSMAR

Note: The data are collected from well-known databases, including CSMAR and CNRDS.

).

3.2. Variables Construction

3.2.1. Firm Performance

Given that Return on Assets (ROA) is the most used financial metric [12,13], we use it to evaluate firm financial performance (FP) [58].

3.2.2. Green Innovation

Green innovation is measured by reference to current related research [8] through two distinct metrics: Total Green Innovation (EnvrPats) is defined as the total number of patent applications for green inventions and green utility model applications, plus one [7], and Green Inventive Patents (EnvrInvPats), which counts only the green invention patent applications plus one. Both measures are log-transformed to normalize the data and facilitate a more meaningful econometric analysis. While patent data, such as total green innovation (EnvrPats) and green inventive patents (EnvrInvPats), offer valuable insights into the technological dimension of green innovation, they have limitations. Not all green innovation efforts are patentable, and many environmental initiatives such as process improvements, certifications, and organizational changes are not captured by patent data. Therefore, we acknowledge that relying solely on patents may understate the broader scope of green innovation.

3.2.3. Channel Test Variable

This study examines the role of financial constraints as a moderating factor in the relationship between green innovation and financial performance [59]. Using the KZ index to measure financial limitations, it finds that companies with greater financial constraints may struggle to invest in green technologies, limiting their positive impact on performance [60]. Conversely, firms with fewer constraints are more likely to invest in green projects, boosting performance.

It also highlights the importance of R&D expenditure as a driver of green innovation, although high R&D spending can lead to short-term financial strain, with returns realized in the long term [61]. R&D intensity, measured as the ratio of R&D to sales, helps assess a firm’s commitment to innovation.

3.2.4. Moderating Variable

This study includes Sustainable Liquidity Reserves, a measure of cash flow, as a moderating variable to evaluate its influence on the relationship between green innovation and firm profitability. Sustainable Liquidity Reserves highlight a firm’s financial flexibility, indicating its ability to fund long-term projects like green innovation without compromising short-term financial stability. Firms with higher Sustainable Liquidity Reserves are better positioned to absorb the upfront costs associated with innovation, allowing them to sustain these efforts even during financial challenges. This variable is calculated as the ratio of operating cash flow to total assets, providing a measure of how effectively a firm generates cash from its core activities relative to its asset base [18,19].

This study uses the Sustainable Development Index to capture ESG performance, with higher scores indicating better performance in sustainability and corporate governance. For Chinese firms, factors such as compliance with China’s Environmental Protection Law and adherence to state-mandated social responsibility standards are critical components of the ESG score [62]. The Sustainable Development Index serves as a moderating variable to examine how strong ESG performance influences the relationship between green innovation and financial outcomes.

3.2.5. Control Variables

Continuous variables are winsorized at the 1st and 99th percentiles to reduce the impact of outliers [63]. Equation (1) incorporates controls for fixed factors related to year, country, and industry. According to the relevant research, the subsequent variables are designated as control variables. This study incorporates numerous essential control variables to assess the effect of green innovation on company performance accurately. The control variables encompass the size of the company, generally quantified by the natural logarithm of total assets, as larger organizations frequently possess enhanced resources and economies of scale that affect innovation results [63]. Leverage, typically quantified as the ratio of total assets to debt, influences a company’s financial risk and its capacity to engage in innovation. Leverage commonly measured as the ratio of total debt total assets, affects a firm’s financial risk and its ability to invest in innovation. A higher level debt can either constrain or enhance firm performance, depending on the firm’s ability to manage financial obligations [64]. R&D expenditures represent a firm’s commitment to innovation and long-term growth. Controlling for R&D (InRD) ensures that the effect of innovation is distinguished from other factors driving firm performance [65]. Sales growth measures a firm’s ability to increase revenue over time. It indicates a company’s performance regarding market expansion and product demand [66]. The company’s age, or the duration from its inception, is considered as an older firm that may possess established systems that influence its ability for innovation [7]. The market-to-book ratio indicates investors’ anticipations regarding a company’s future growth and performance. It assesses the market’s valuation of a firm relative to its book value. These controls help isolate the specific impact of green innovation on perform by accounting for other factors that could influence the results.

3.3. Econometric Modeling

To empirically investigate the relationship between green innovation and firm performance, this study employs the following econometric models:

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(1)

This equation quantifies the impact of green innovation ${\mathrm{G}\mathrm{I}}_{\mathrm{i},\mathrm{t}}$. The coefficient ${\beta }_1$ quantifies the impact of green innovation on performance metrics like Return on Assets (ROA). The model includes control variables ${con}_{i,t}$, fixed effects ${\alpha }_{i,t}$ to account for firm-specific unobserved factors, and time effects ${\gamma }_{i,t}$ to account for time-specific shocks.

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(2)

$${R\&D}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(3)

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{\beta }_2{R\&D}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(4)

Equation (2) examines the GI influence the FP, and (3) examines how green innovation influences a firm’s R&D expenditure, while Equation (4) investigates the combined effect of green innovation and R&D on financial performance. Together, they analyze the direct and indirect impact of green innovation on financial performance, with R&D acting as a channel in the relationship.

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(5)

$${FC}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(6)

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{\beta }_2{FC}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(7)

This equation assesses how both green innovation and financial constraints jointly affect firm performance. The coefficient β₁ captures the direct impact of green innovation on financial performance, while β₂ represents the effect of financial constraints on firm performance, reflecting whether firms experiencing higher financial constraints due to green innovation face adverse financial outcomes. A significant β₂ would suggest that financial constraints as a channel the relationship between green innovation and financial performance, implying that the financial strain or flexibility related to green innovation activities can influence overall firm performance. Similar to Equations (8) and (9), control variables and fixed effects are included to ensure the robustness of the results.

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{\beta }_2{CF}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(8)

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{\beta }_2{CF}_{i,t}+{\beta }_3\left({GI}_{i,t}\ast {CF}_{i,t}\right)+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(9)

These equations introduce a moderation model where ${CF}_{i,t}$ represents cash flow as a moderator. The second equation includes the interaction term (${GI}_{i,t}$ × ${CF}_{i,t}$), which shows how the effect of green innovation on firm performance changes depending on the level of cash flow.

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{\beta }_2{ESG}_{i,t}+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(10)

$${FP}_{i.t}={\beta }_0+{\beta }_1{GI}_{i,t}+{\beta }_2{ESG}_{i,t}+{\beta }_3\left({GI}_{i,t}*{ESG}_{i,t}\right)+{con}_{i,t}+{\alpha }_{i,t}+{\gamma }_{i,t}+{\epsilon }_{i,t}$$

(11)

These models explore not only the independent contributions of green innovation and ESG performance to firm performance but also how ESG performance can moderate the effectiveness of green innovation initiatives. Equation (10) assesses the direct effects, while Equation (11) investigates whether firms with strong ESG performance derive additional financial benefits from their green innovation efforts due to the interaction between these two factors. This approach provides a deeper understanding of the role of ESG in amplifying or mitigating the financial impact of sustainable innovation.

We conducted a comprehensive empirical examination to assess the relationship between green innovation and firm financial performance in high-pollution industries. As a preliminary step, descriptive statistics and correlation analysis were employed to explore data characteristics and inter-variable relationships. Ordinary Least Squares (OLS) regression was used as a baseline model to provide initial estimates of the associations, offering a benchmark for comparison with more advanced techniques. To control for unobserved firm-level heterogeneity and time effects, we applied fixed-effects regression with robust standard errors, addressing potential concerns of heteroscedasticity and serial correlation.

Given the potential for endogeneity commonly encountered in studies of innovation and corporate performance, we implemented several identification strategies. First, we employed a two-stage least squares (2SLS) estimation using digital transformation as an instrumental variable, leveraging its theoretical link to innovation without a direct impact on short-term financial outcomes [67,68]. Second, we applied the system Generalized Method of Moments (System GMM) estimator [40] to address simultaneity bias, dynamic panel effects, and endogeneity concerns, incorporating lagged dependent variables as internal instruments.

To capture nonlinear dynamics, quadratic terms of green innovation were introduced, revealing a U-shaped relationship with financial performance. Further, mediation and moderation analyses were conducted to identify key internal mechanisms (R&D expenditure and financial constraints) and firm-level conditions (cash flow and ESG performance) that shape the green innovation–performance relationship. Robustness checks using alternative proxies such as Return on Equity (ROE) and the total number of green patents confirmed the consistency of the findings. This multimethod strategy enhances the validity and reliability of the empirical results, providing a nuanced understanding of how green innovation influences firm financial performance under varying internal and external conditions.

4. Results and Empirical Analysis

4.1. Summary Statistics

Table 1. Descriptive statistics.

Variables	N	Mean	SD	Min	p25	Median	p75	Max
ROA	35,784	2.539	3.294	−7.708	0.759	2.182	4.139	13.463
ROE	35,773	4.124	6.289	−23.401	1.524	4.053	7.081	22.098
EnvrPats	35,784	0.327	0.731	0	0	0	0	3.497
EnvrInvPats	35,784	0.219	0.577	0	0	0	0	3.045
green N	35,784	1.115	9.134	0	0	0	0	559
lnRD	35,784	0	1.000	−0.57	−0.57	−0.57	−0.57	2.301
LEV	35,784	0.423	0.205	0.054	0.257	0.416	0.575	0.882
age	35,784	10.526	7.448	1	4	9	16	32.00
MB	35,784	2.075	1.289	0.865	1.276	1.662	2.366	8.314
growth	35,747	0.37	0.999	−0.72	−0.035	0.126	0.406	7.047
size	35,784	22.183	1.296	19.876	21.243	21.989	22.917	26.243
KZ	35,784	1.098	2.428	−5.886	−0.342	1.332	2.716	6.671

Note: This table provides the descriptive statistics for the key variables used in this study. ROA and ROE reflect firm profitability. EnvrPats and EnvrInvPats measure the extent of environmental innovation. Green N indicates the number of sustainability initiatives. lnRD, LEV, Age, MB, Growth, Size, and KZ represent control variables related to research and development, financial leverage, firm age, market valuation, sales growth, firm size, and financial constraints, respectively.

summarizes the descriptive statistics of key variables such as ROA, ROE, green innovation (GI), and leverage, showcasing the variability in firm performance and innovation activities [8]. The average ROA of 0.0374 and ROE of 0.0841 indicate consistent profitability, albeit with noticeable differences in financial performance across firms [8]. The mean value of green innovation, measured by green patents (0.4195), suggests that while some firms are actively engaging in eco-friendly innovation, others are not. The leverage ratio, with an average of 0.4252, indicates a moderate level of debt utilization, reflecting diverse capital structure strategies among firms. These results align with previous studies supporting Hypothesis 1, which posits that both green innovation and financial leverage can positively impact firm performance when effectively managed [32], in accordance with the Resource-Based View and Porter’s Hypothesis. The descriptive statistics provide empirical evidence supporting the proposed relationships in this study. The results align with the basis that a firm’s financial and resource limitations substantially affect the relationship between green innovation and economic performance. Companies with a robust financial position and increased R&D expenditure are more capable of absorbing the expenditures linked to green innovation, in accordance with established literature on sustainability and corporate performance [47]. The data indicate a positive relationship between green patents and ROA, implying that the initial expenditures associated with green inventions may not produce timely financial returns due to substantial costs and prolonged development timelines.

4.2. Correlation Matrix

The empirical examination of the correlation matrix in

Table 2. Correlation Matrix.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)
(1) ROA	1.000
(2) ROE	0.889	1.000
(3) EnvrPats	0.021	0.046	1.000
(4) EnvrInvPats	0.020	0.043	0.922	1.000
(5) green_N	0.010	0.033	0.431	0.484	1.000
(6) lnRD	−0.047	−0.043	0.093	0.115	0.067	1.000
(7) LEV	−0.341	−0.131	0.071	0.062	0.051	0.005	1.000
(8) age	−0.170	−0.084	−0.071	−0.047	0.017	0.061	0.345	1.000
(9) MB	0.222	0.113	−0.047	−0.033	−0.029	0.043	−0.282	−0.063	1.000
(10) growth	−0.081	−0.042	−0.015	−0.004	−0.008	0.023	0.064	0.065	0.041	1.000
(11) size	−0.012	0.116	0.177	0.181	0.156	0.120	0.504	0.387	−0.398	−0.008	1.000
(12) KZ	−0.576	−0.417	−0.020	−0.021	−0.012	0.038	0.616	0.259	−0.034	0.069	0.121	1.000

Note: This table presents the correlation coefficients among key variables, including financial performance metrics (ROA, ROE), green innovation indicators (EnvrPats, EnvrInvPats), and control variables such as leverage (LEV), firm size, and financial constraints (KZ). Significant correlations suggest potential interactions among these variables, which are essential for understanding the relationships explored in this study.

offers critical insights into the interrelation of financial performance, green innovation, and business attributes. Return on Assets (ROA) and Return on Equity (ROE) exhibit a robust positive correlation (0.889), indicating a close relationship between profitability indicators. However, the correlations between green innovation variables, such as environmental patents (EnvrPats) and environmentally innovative patents (EnvrInvPats), with ROA (0.021 and 0.020) and ROE (0.046 and 0.043) are weak, indicating that green innovation initiatives have a constrained direct influence on financial results. This may result from the substantial initial expenditures and the long duration of investments in green innovations [69]. Moreover, leverage (LEV) exhibits a negative connection with both ROA (−0.341) and ROE (−0.131), indicating that firms with elevated debt levels generally experience lowered profitability. The Kaplan–Zingales index (KZ) has a strong negative correlation with ROA (−0.576) and ROE (−0.417), highlighting that firms facing greater financial constraints are less profitable. The positive correlation between firm size and green innovation variables suggests that larger firms are more likely to engage in green innovation, likely due to greater resource availability. Overall, the matrix reveals that while firm size and financial health play crucial roles in green innovation, the immediate financial benefits are not strongly apparent.

4.3. Estimation Results

4.3.1. Baseline Regression

The baseline regression results presented in

Table 3. High-Dimension Baseline Regression: OLS and Fixed Effect Results.

	(1)	(2)	(3)	(4)
VARIABLES	ROA (T + 1)	ROA (T + 1)	ROA (T + 1)	ROA (T + 1)
EnvrPats	−0.0863 ***		−0.0739 ***
	(0.0240)		(0.0072)
EnvrInvPats		−0.0961 ***		−0.0647 ***
		(0.0254)		(0.0130)
LEV	−4.9646 ***	−6.2430 ***	−2.1325 **	−2.1334 **
	(0.1024)	(0.0974)	(0.6453)	(0.6446)
MB	0.5825 ***	0.6165 ***	0.4012 ***	0.4010 ***
	(0.0150)	(0.0193)	(0.0246)	(0.0245)
Growth	−0.1607 ***	−0.1781 ***	0.0365	0.0365
	(0.0171)	(0.0173)	(0.0224)	(0.0223)
Size	0.6433 ***	0.8616 ***	−0.6577 ***	−0.6584 ***
	(0.0174)	(0.0162)	(0.1275)	(0.1277)
AGE	−0.0487 ***	−0.0658 ***
	(0.0027)	(0.0024)
Constant	−10.3267 ***	−14.4298 ***	17.1329 ***	17.1384 ***
	(0.3752)	(0.3466)	(2.9846)	(2.9876)
Observations	30,501	35,747	30,108	30,108
R-squared	0.1481	0.2116	0.5580	0.5580

Note: Standard errors are shown in parentheses. Statistical significance levels are indicated as follows: *** denotes p < 0.01 (highly significant), and ** denotes p < 0.05 (moderately significant).

investigate the effect of green innovation on firm financial performance, measured by Return on Assets (ROA) [70,71]. Green innovation is proxied using environmental patents (EnvrPats) and environmentally innovative patents (EnvrInvPats) [72]. Across all model specifications, the coefficients of both indicators are negative and statistically significant at the 1% level [73,74]. These results indicate that green innovation is associated with a short-term decline in firm profitability.

This finding provides strong support for Hypothesis 1, which posits that green innovation exerts a negative impact on short-term financial performance. The observed decline likely reflects the high upfront costs, extended commercialization periods, and delayed financial returns associated with the implementation of green technologies. These baseline results underscore the financial trade-offs firms may face when engaging in green innovation and justify the need for further analysis into underlying mechanisms and moderating factors.

4.3.2. Mechanism Analysis

The mechanism analysis investigates how green innovation impacts firm performance (ROA) through two internal channels: R&D expenditure and financial constraints. As shown in

Table 4. Channel test using RD.

	(3)	(1)	(2)		(5)	(6)
VARIABLES	ROA	RD	ROA	ROA	RD	ROA
EnvrPats	−0.0739 ***	0.0314 **	−0.0708 ***
	(0.0072)	(0.0097)	(0.0070)
EnvrInvPats				−0.0647 ***	0.0390 **	−0.0595 ***
				(0.0130)	(0.0135)	(0.0128)
lnRD			−0.1420 ***			−0.1421 ***
			(0.0173)			(0.0173)
LEV	−2.1325 **	−0.2089 ***	−2.1439 **	−2.1334 **	−0.2085 ***	−2.1448 **
	(0.6453)	(0.0351)	(0.6390)	(0.6446)	(0.0349)	(0.6383)
MB	0.4012 ***	0.0163 ***	0.4023 ***	0.4010 ***	0.0163 ***	0.4020 ***
	(0.0246)	(0.0028)	(0.0242)	(0.0245)	(0.0028)	(0.0241)
growth	0.0365	0.0079	0.0375	0.0365	0.0079	0.0376
	(0.0224)	(0.0076)	(0.0225)	(0.0223)	(0.0076)	(0.0225)
size	−0.6577 ***	0.2071 ***	−0.6265 ***	−0.6584 ***	0.2070 ***	−0.6272 ***
	(0.1275)	(0.0200)	(0.1225)	(0.1277)	(0.0200)	(0.1226)
Constant	17.1329 ***	−4.5148 ***	16.4424 ***	17.1384 ***	−4.5097 ***	16.4487 ***
	(2.9846)	(0.4415)	(2.8678)	(2.9876)	(0.4406)	(2.8709)
Sobel test	−11.165 ***			−12.762 ***
Observations	30,108	30,108	30,108	30,108	30,108	30,108
R-squared	0.5580	0.7255	0.5586	0.5580	0.7255	0.5585

Note: Standard errors are in parentheses. ***, ** indicate significance at the 1% and 5% levels, respectively, showing the statistical reliability of the coefficients related to green innovation and firm performance variables.

green innovation significantly increases R&D spending, which in turn negatively affects ROA, indicating that the short-term financial burden of innovation activities reduces profitability. Similarly,

Table 5. Channel test using KZ.

	(1)	(2)	(3)	(4)	(5)	(6)
VARIABLES	Roa1	KZ	ROA	Roa2	KZ	ROA
EnvrPats	−0.0739 ***	0.0575 ***	−0.0658 ***
	(0.0072)	(0.0036)	(0.0051)
EnvrInvPats				−0.0647 ***	0.0666 ***	−0.0567 ***
				(0.0130)	(0.0068)	(0.0111)
KZ			−0.4441 ***			−0.4441 ***
			(0.0188)			(0.0187)
LEV	−2.1325 **	4.6520 ***	1.2082 **	−2.1334 **	4.6527 ***	1.2078 **
	(0.6453)	(0.2804)	(0.4343)	(0.6446)	(0.2805)	(0.4336)
MB	0.4012 ***	0.0938 ***	0.5232 ***	0.4010 ***	0.0939 ***	0.5230 ***
	(0.0246)	(0.0167)	(0.0346)	(0.0245)	(0.0167)	(0.0344)
growth	0.0365	−0.0104	0.0203	0.0365	−0.0104	0.0203
	(0.0224)	(0.0079)	(0.0271)	(0.0223)	(0.0079)	(0.0270)
size	−0.6577 ***	0.1663 ***	−0.7674 ***	−0.6584 ***	0.1662 ***	−0.7680 ***
	(0.1275)	(0.0295)	(0.1373)	(0.1277)	(0.0295)	(0.1376)
Constant	17.1329 ***	−4.6949 ***	18.3644 ***	17.1384 ***	−4.6886 ***	18.3699 ***
	(2.9846)	(0.6635)	(3.1097)	(2.9876)	(0.6623)	(3.1142)
Sobel test	3.735 ***			3.081 ***
Observations	30,108	30,108	30,108	30,108	30,108	30,108
R-squared	0.5580	0.6631	0.5888	0.5580	0.6631	0.5887

Note: Standard errors are in parentheses. ***, ** denote significance at the 1% and 5% levels, respectively. The KZ index measures financial constraints, with negative coefficients on ROA indicating the impact of green innovation and financial variables on firm performance.

reveals that green innovation heightens financial constraints, as measured by the KZ index, which also contributes to lower ROA. These results confirm that while green innovation holds long-term value, it imposes immediate financial strain through increased investment and limited liquidity. The significant Sobel test results further validate the indirect effects, highlighting the importance of financial planning and resource management in sustaining green innovation efforts without compromising near-term performance.

Channel Test Analysis

The channel test results in Table 4 investigate the mediating role of R&D expenditure in the relationship between green innovation and firm financial performance. In Column (2), environmental patents (EnvrPats) are positively associated with R&D intensity, indicating that firms engaged in green innovation allocate greater resources to research and development [75]. However, as shown in Column (3), this increased R&D spending is associated with a decline in ROA, suggesting that the short-term financial burden of innovation investment may outweigh immediate returns. Similar results are observed in Columns (5) and (6) when environmentally innovative patents (EnvrInvPats) are used, reinforcing the role of R&D as a financial pressure point during the early stages of green innovation.

In Table 5, financial constraints are assessed as a second mediating channel using the KZ index [76]. The results in Column (2) demonstrate a positive association between green innovation and financial constraints, implying that environmental innovation intensifies internal financing pressures. Column (3) shows that higher financial constraints negatively affect ROA, confirming that restricted liquidity can impair short-term profitability. Parallel effects using EnvrInvPats in Columns (5) and (6) further validate this pathway. The Sobel test results confirm the statistical significance of both mediation effects.

The empirical evidence indicates that green innovation influences short-term financial performance through two distinct internal mechanisms: greater investment in R&D and increased financial constraints [67,68]. These findings reflect the underlying cost structures and liquidity pressures that firms face when pursuing environmentally focused strategies. The results emphasize the financial trade-offs associated with early-stage green innovation, particularly in capital-intensive and heavily regulated industries. This underscores the importance of financial preparedness and strategic resource management, enabling firms to absorb transitional costs while positioning themselves for long-term performance benefits.

4.3.3. Moderation Analysis

Table 6. Moderation Test.

	(1)	(2)	(4)	(3)
VARIABLES	ROA (T + 1)	ROA (T + 1)	ROA (T + 1)	ROA (T + 1)
EnvrPats	0.7225 *		1.2386 **
	(0.3514)		(0.4018)
EnvrInvPats		0.9039 **		2.2307 ***
		(0.3121)		(0.4477)
ESG	0.0213 ***	0.0205 ***
	(0.0050)	(0.0045)
CFO			0.7498 ***	0.6893 ***
			(0.0373)	(0.0509)
EnvrPats * ESG	−0.0106 *
	(0.0047)
EnvrInvPats * ESG		−0.0129 **
		(0.0042)
EnvrPats * CFO			−0.0643 **
			(0.0198)
EnvrInvPats * CFO				−0.1115 ***
				(0.0222)
LEV	−2.1600 **	−2.1617 **		−0.9365
	(0.5530)	(0.5514)		(0.6208)
AGE	−0.0352	−0.0355
	(0.0406)	(0.0407)
MB	0.3587 ***	0.3585 ***	0.3948 ***	0.4002 ***
	(0.0115)	(0.0115)	(0.0226)	(0.0238)
Growth	0.0477 *	0.0477 *	0.0279	0.0308
	(0.0231)	(0.0230)	(0.0240)	(0.0254)
Size	−0.6596 ***	−0.6605 ***	−1.4597 ***	−1.3283 ***
	(0.1471)	(0.1474)	(0.1734)	(0.1252)
Constant	16.0842 ***	16.1575 ***	18.9842 ***	17.6564 ***
	(3.1894)	(3.1744)	(3.4839)	(2.8994)
Observations	30,108	30,108	30,108	30,108
R-squared	0.5478	0.5478	0.5648	0.5655

Note: This table presents the moderation analysis results examining the effects of cash flow (CFO) and ESG scores on the relationship between green innovation and firm performance (ROA). Interaction terms indicate diminishing returns from increased CFO and ESG. Standard errors are in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the results of the moderation study examining the influence of cash flow (CFO) and ESG performance on the relationship between green innovation and business performance (ROA). The baseline results show that green innovation, measured by environmental patents, positively contributes to firm performance. However, the interaction terms reveal a more complex picture: as CFO and ESG performance increase, the positive impact of green innovation on ROA diminishes. This suggests that while green innovation initially enhances performance, increasing cash flow and ESG efforts may introduce additional costs, reducing short-term financial gains. The findings highlight the challenges firms face in balancing the financial and ethical costs of green innovation, emphasizing the need for strategic resource management. The models explain a substantial portion of the variation in ROA (R-squared values between 54.8% and 56.5%), and the inclusion of fixed and year effects strengthens the robustness of the results.

4.4. Heterogeneity Analysis

4.4.1. Assessing Green Innovation’s Effectiveness in Varying Pollution Context: A Heterogeneity Approach

Table 7. Heterogeneity Analysis by Industry Pollution Level.

	(1)	(2)
Variables	High Pollution	Low Pollution
EnvrPats	−0.1677 ***
	(0.0183)
EnvrInvPats		−0.1650 ***
		(0.0182)
pollution	0.4262	0.4421
	(0.3950)	(0.3985)
EnvrPats × Pollution	0.3724 ***
	(0.0215)
EnvrInvPats × Pollution		0.4167 ***
		(0.0220)
LEV	−2.0942 **	−2.0991 **
	(0.6165)	(0.6162)
MB	0.4021 ***	0.4020 ***
	(0.0250)	(0.0249)
Growth	0.0366	0.0366
	(0.0224)	(0.0223)
Size	−0.6536 ***	−0.6552 ***
	(0.1287)	(0.1289)
Constant	16.9294 ***	16.9512 ***
	(2.9566)	(2.9601)
Observations	30,108	30,108
R-squared	0.5587	0.5586

Note: Standard errors are in parentheses. *** p < 0.01, ** p < 0.05. Green innovation initially decreases ROA due to high costs, but firms in high-pollution industries see a positive effect from regulatory compliance and efficiency gains. Results are robust, accounting for firm-specific and time-related factors.

examines how the impact of green innovation on firm performance (measured by ROA) varies between high- and low-pollution industries. The negative coefficients for environmental patents (EnvrPats = −0.168) and environmentally innovative patents (EnvrInvPats = −0.165) suggest that green innovation generally leads to a short-term decline in ROA due to high initial costs and delayed returns. However, the positive and significant interaction terms with the pollution dummy variable (EnvrPats × pollution = 0.372 and EnvrInvPats × pollution = 0.417) indicate that this negative impact is mitigated or even reversed in high-pollution industries. These findings imply that firms in high-pollution sectors benefit more from green innovation [77], as it helps them comply with stricter environmental regulations and achieve operational efficiencies, reducing regulatory risks and costs [78]. In contrast, firms in low-pollution industries, facing less regulatory pressure and weaker incentives, do not experience the same short-term benefits and may see a negative impact on ROA. This highlights the strategic importance of green innovation in high-pollution industries, where it can improve market position and financial performance by addressing both regulatory and operational challenges.

4.4.2. Shock Analysis—Pre- and Post-COVID-19 Test

Table 8. Heterogeneity Analysis by COVID-19 Shocks.

VARIABLES	>2020	<2020	>2020	<2020
EnvrPats	0.1542 ***	0.1093 ***
	−0.0588	−0.0413
EnvrInvPats			0.1797 **	0.1163 **
			−0.0707	−0.052
LEV	−5.9617 ***	−6.4961 ***	−5.9676 ***	−6.4953 ***
	−0.1922	−0.1194	−0.1922	−0.1194
AGE	−0.0336 ***	−0.0474 ***	−0.0336 ***	−0.0474 ***
	−0.0043	−0.0033	−0.0043	−0.0033
MB	0.7888 ***	0.5822 ***	0.7902 ***	0.5822 ***
	−0.0263	−0.0182	−0.0263	−0.0182
Growth	−0.3194 ***	−0.1329 ***	−0.3190 ***	−0.1330 ***
	−0.0393	−0.0193	−0.0393	−0.0193
Size	0.9285 ***	0.8638 ***	0.9302 ***	0.8636 ***
	−0.0304	−0.0204	−0.0304	−0.0204
Constant	−17.3048 ***	−14.4209 ***	−17.3462 ***	−14.4171 ***
	−0.6559	−0.4472	−0.6558	−0.4473
Observations	7054	20,334	7054	20,334
R-squared	0.2664	0.2037	0.2663	0.2036

Note: Standard errors are reported in parentheses. *** p < 0.01, ** p < 0.05. analyzes the impact of green innovation on firm performance, measured by ROA, before and after the COVID-19 pandemic.

analyzes the impact of green innovation on firm performance, measured by ROA, before and after the COVID-19 pandemic. The results show that green innovation, indicated by changes in environmental patents (EnvrPats) and environmentally innovative patents (EnvrInvPats), has a positive and significant effect on firm performance in both periods. However, the effect is stronger post-COVID-19, with higher coefficients (0.1542 and 0.1797, respectively) compared with the pre-pandemic period (0.1093 and 0.1163). This suggests that the importance of green innovation has increased during the pandemic, possibly due to heightened regulatory demands and the need for resilience. The analysis also reveals that higher leverage and older firm age are associated with lower ROA in both periods, indicating that financial constraints and legacy systems may hinder performance. The negative impact of leverage was more pronounced pre-COVID-19, suggesting that firms with higher debt faced greater challenges in funding green innovation before the pandemic. Larger firms with higher market-to-book ratios performed better, while rapid growth negatively affected short-term profitability. The improved R-squared value post-COVID-19 (26.64% vs. 20.37%) indicates that the pandemic context has enhanced the model’s explanatory power. Overall, the findings highlight the amplified strategic value of green innovation for firm performance during the pandemic.

4.5. Addressing Endogeneity Issue

4.5.1. Dynamic Panel Two-Stage Generalized Method of Moments: Addressing Endogeneity Concerns

In order to reduce potential endogeneity between green innovation and financial performance, the two-stage least square (2SLS) strategy is employed by using the ivreg method, integrating the work in [40], with digital transformation (DT) as an instrumental variable (IV) [79]. In

Table 9. 2SLS.

	2SLS	2SLS
VARIABLES	ROA (T + 1)	ROA (T + 1)
Instrumental Variable = Digital Transformation
EnvrPats	0.361 ***
	(0.087)
EnvrInvPats		0.442 ***
		(0.107)
LEV	−6.453 ***	−6.481 ***
	(0.095)	(0.095)
Age	−0.068 ***	−0.067 ***
	(0.003)	(0.003)
MB	0.609 ***	0.611 ***
	(0.014)	(0.014)
Growth	−0.200 ***	−0.198 ***
	(0.017)	(0.017)
Size	0.933 ***	0.934 ***
	(0.020)	(0.020)
Constant	−15.864 ***	−15.905 ***
	(0.418)	(0.423)
Observations	33,068	33,068
Firm FE	Yes	Yes
Year FE	Yes	Yes
LM Stat	2319.51 **	2692.22 **
Wald test	2529.10 **	2978.91 **
Adjusted R2	0.210	0.211

Note: Standard errors are reported in parentheses. *** p < 0.01, ** p < 0.05. This table reports the second-stage results of the two-stage least squares (2SLS) estimation, assessing the impact of green innovation on firm performance (ROA). Digital transformation is used as an instrumental variable to address potential endogeneity. EnvrPats and EnvrInvPats represent environmental patents and environmentally innovative patents, respectively. Firm and year fixed effects are included to control for unobserved heterogeneity. The LM statistic confirms the relevance of the instrument, and the Wald test supports the joint significance of the repressor’s. The results suggest a positive and significant effect of green innovation on ROA when endogeneity is accounted for.

, the first stage reveals a significant negative relationship between DT and green innovation (EnvrPats and EnvrInvPats), indicating that firms focusing on digital transformation may initially allocate fewer resources to green initiatives. This suggests a trade-off in resource allocation between digital and green projects [80]. In the second stage, after addressing endogeneity, green innovations exhibit a positive and significant effect on financial performance (ROA), with coefficients of −0.361 for EnvrPats and −0.442 for EnvrInvPats. This confirms that, despite initial resource constraints, investing in green innovation ultimately enhances financial performance. The validity of DT as an IV is supported by strong LM and Wald test results, indicating that DT influences ROA only through its impact on green innovation. This analysis emphasizes the strategic significance of green innovation in attaining sustainable financial returns.

4.5.2. Dynamic Panel of Generalized Method of Moments: Endogeneity Issues

The Generalized Method of Moments (GMM) estimation results displayed in

Table 10. Generalized Method of Moments (GMM) Estimated Results.

	(1)	(2)
VARIABLES	GMM	GMM
L.ROA	0.5925 ***	0.9327 ***
	(0.0187)	(0.0223)
EnvrPats	−0.0374 *
	(0.0195)
EnvrInvPats		−0.0824 ***
		(0.0308)
LEV	−1.8963 ***	0.1456
	(0.1691)	(0.1897)
Age	−0.0088 ***	0.0028
	(0.0028)	(0.0025)
MB	0.3748 ***	0.2056 ***
	(0.0256)	(0.0274)
Growth	−0.0312 *	0.0340
	(0.0175)	(0.0244)
Size	0.2968 ***	0.0676 ***
	(0.0248)	(0.0219)
Constant	−6.6426 ***	−2.4149 ***
	(0.4817)	(0.4347)
Observations	30,507	17,446
Number of code	3924	2721
AR1	0.002	0.004
AR2	0.789	0.654
Sargan	0.231	0.572
Hansen	0.124	0.343

Note: Standard errors are reported in parentheses. *** p < 0.01, * p < 0.1. The table presents the results of the Generalized Method of Moments (GMM) estimation, where L.ROA is the lagged return on assets, EnvrPats represents environmental patents, and EnvrInvPats denotes environmentally innovative patents. The AR1 and AR2 statistics test for autocorrelation in the error terms, with significant AR1 and nonsignificant AR2 values indicating valid moment conditions. The Sargan and Hansen tests assess the validity of the instruments, with high p-values indicating no over-identification and confirming the exogeneity of the instruments used.

offer significant insights into the relationship between green innovations and financial performance. The significant and positive coefficients for the lagged ROA variable (0.5925 and 0.9327) in both models indicate strong persistence in firm performance, suggesting that past profitability is a key predictor of current financial success. In contrast, the coefficients for green innovation variables are negative, with environmental patents (EnvrPats) showing a marginally significant negative impact on ROA (−0.0374), and environmentally innovative patents (EnvrInvPats) displaying a more substantial negative effect (−0.0824). These results imply that while green innovation is crucial for long-term sustainability, it may initially entail financial costs that can reduce short-term profitability. Control variables explain the underlying dynamics. The initial model shows that leverage has a substantial negative influence on ROA, suggesting that high debt levels may negatively impact financial performance. The market-to-book ratio and firm size exert a positive and considerable influence on ROA, indicating that companies with better growth potential and grander operational scale achieve financial advantages. Diagnostic assessments, encompassing AR1, AR2, Sargan, and Hansen tests, validate the model’s resilience and precision. The substantial findings of the AR1 test demonstrate first-order autocorrelation. However, the nonsignificant outcomes of the AR2 test affirm the lack of second-order autocorrelation, so validating the employed moment circumstances. Moreover, the high p-values from the Sargan and Hansen tests demonstrate that the instruments are not over-identified, confirming their exogeneity and appropriateness in addressing endogeneity concerns. Overall, the findings suggest that although green innovation may temporarily reduce financial performance due to its associated costs, it remains an essential strategic investment for sustainable growth. The results emphasize the trade-offs firms face between pursuing sustainability and maintaining immediate profitability. The robustness of these conclusions is further reinforced by the diagnostic tests, validating the use of digital transformation as an effective instrumental variable to address endogeneity in this context.

4.6. Robustness Check

4.6.1. Proxy Replacement

To ensure the robustness of the findings, we re-estimate the models using Return on Equity (ROE) as an alternative proxy for financial performance, replacing the original ROA measure. As shown in

Table 11. Substitute ROA with ROE.

	(1)	(2)	(3)	(4)
VARIABLES	ROE (T + 1)	ROE (T + 1)	ROA (T + 1)	ROE (T + 1)
EnvrPats	−0.1556 ***
	(0.0361)
EnvrInvPats		−0.1408 ***
		(0.0197)
Green_N			−0.0034 ***	−0.0121 ***
			(0.0006)	(0.0024)
LEV	−0.4053	−0.4072	−2.1352 **	−0.4140
	(1.3215)	(1.3206)	(0.6441)	(1.3219)
MB	0.7213 ***	0.7209 ***	0.4009 ***	0.7212 ***
	(0.0439)	(0.0437)	(0.0246)	(0.0440)
Growth	0.1323	0.1323 *	0.0366	0.1325 *
	(0.0657)	(0.0656)	(0.0222)	(0.0653)
Size	−1.3129 ***	−1.3141 ***	−0.6601 ***	−1.3168 ***
	(0.2016)	(0.2016)	(0.1276)	(0.2015)
Constant	31.8742 ***	31.8825 ***	17.1672 ***	31.9279 ***
	(4.8570)	(4.8562)	(2.9870)	(4.8525)
Observations	30,102	30,102	30,108	30,102
R-squared	0.4503	0.4503	0.5580	0.4503

Note: Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. Replacing ROA with ROE confirms that green innovation negatively impacts short-term financial performance. This effect persists across different metrics, highlighting the initial costs of green investments. Firm and year fixed effects control for unobserved factors.

, the results remain consistent, with green innovation measured by both environmental patents and environmentally innovative patents showing a negative and significant relationship with short-term financial performance [81]. This indicates that the financial burden of green innovation persists regardless of whether profitability is measured via asset efficiency (ROA) or equity returns (ROE).

Additional robustness tests using the number of green patents (Green_N) as an alternative proxy for green innovation further confirm these patterns. The consistent negative results across both ROA and ROE models suggest that green innovation incurs substantial short-term financial costs, likely due to high upfront investments and delayed returns. By incorporating firm and year fixed effects, along with control variables, the models account for potential unobserved heterogeneity and time-specific effects. Overall, these robustness checks reinforce the validity of the main findings and underline the immediate financial trade-offs firms face when adopting green innovation strategies.

4.6.2. Nonlinear or Converse U-Turn and Long-Term Effect Including the Quadric Term

Table 12. Estimated Nonlinear or Converse U-turn and Long-Term Effect Results.

	(1)	(2)	(3)	(4)
VARIABLES	Long-Term ROA (T + 1)	Long-Term ROA (T + 1)	Nonlinear or Converse U Turn ROA (T + 1)	Nonlinear or Converse U Turn ROA (T + 1)
EnvrPats	−0.1302 **
	(0.0350)
EnvrInvPats		−0.1831 ***
		(0.0446)
Green_N			−0.0034 ***	−0.0076 ***
			(0.0006)	(0.0008)
GreenN2				0.0130 ***
				(0.0028)
LEV	0.0106	0.0094	−2.1352 **	−2.1323 **
	(0.5814)	(0.5818)	(0.6441)	(0.6438)
MB	0.0602	0.0605	0.4009 ***	0.4009 ***
	(0.0453)	(0.0453)	(0.0246)	(0.0246)
Growth	0.0495 ***	0.0493 ***	0.0366	0.0366
	(0.0115)	(0.0115)	(0.0222)	(0.0222)
Size	−1.2732 ***	−1.2709 ***	−0.6601 ***	−0.6598 ***
	(0.1717)	(0.1712)	(0.1276)	(0.1274)
Constant	30.4143 ***	30.3604 ***	17.1672 ***	17.1625 ***
	(3.9865)	(3.9759)	(2.9870)	(2.9830)
Observations	26,404	26,404	30,108	30,108
R-squared	0.5593	0.5593	0.5580	0.5580

Notes: Standard errors in parentheses. *** p < 0.01, ** p < 0.05, The dependent variable is ROA, with firm and year fixed effects included. EnvrPats and EnvrInvPats represent environmental patents. green_N and greenN2 capture the linear and quadratic effects of green patents. The negative coefficient for green_N suggests initial green innovation reduces ROA due to high costs. The positive coefficient for greenN2 indicates that this negative impact diminishes at higher levels, suggesting a threshold effect where continued investment may eventually improve financial performance.

explores the long-term and nonlinear relationship between green innovation and firm performance, focusing on how environmental and inventive environmental patents affect profitability [56]. The results indicate that green innovation is initially associated with a decline in financial performance (ROA), reflecting the short-term costs and delayed returns typical of sustainability investments.

To capture the nonlinear dynamics, the model includes both linear and quadratic terms of green innovation (Green_N and GreenN2). The negative linear term suggests that initial increases in green innovation reduce ROA, while the positive quadratic term indicates that the negative impact diminishes at higher levels of investment [82]. This pattern implies a U-shaped relationship, suggesting that once firms reach a critical mass of green innovation activity, they begin to experience improved financial outcomes.

These findings underscore the importance of a long-term strategic commitment to green innovation. While early-stage investments may pressure short-term profitability, sustained and scaled efforts can eventually lead to positive financial returns, reinforcing the value of green innovation as a path to sustainable growth.

5. Discussion and Implication

5.1. Discussion

This study provides new empirical insights into the complex relationship between green innovation and firm financial performance, focusing on Chinese A-share listed firms operating in high-pollution industries. The baseline results reveal that green innovation is associated with a decline in short-term profitability, which is consistent with prior research suggesting that the early stages of adopting environmentally innovative practices are financially burdensome due to high R&D investments, regulatory compliance costs, and delayed financial returns [7,8]. These outcomes are particularly evident in firms undergoing the initial phases of green transformation, where the economic benefits of sustainability investments may not immediately materialize.

Further analysis employing a nonlinear specification identifies a U-shaped relationship between green innovation and financial performance [19]. This finding indicates that while green innovation initially reduces firm profitability, the negative effects diminish as innovation intensity increases, eventually leading to improved financial outcomes. This threshold effect supports the view that sustained and scaled innovation efforts enable firms to reap long-term strategic benefits, including cost efficiencies, improved stakeholder perception, and regulatory compliance.

The moderating role of internal firm capabilities further nuances the green innovation–performance link. Specifically, financial flexibility (proxied by cash flow) and ESG performance significantly influence the magnitude of green innovation’s financial impact [83]. Firms with greater liquidity and strong ESG standing are better positioned to manage the costs associated with sustainability initiatives. However, the marginal benefit of these resources appears to decline beyond a certain level, suggesting that the effectiveness of internal capabilities is not linear. This finding extends prior work that highlights the importance of firm-specific resources in enabling the execution of environmental strategies [84,85].

Mediation analysis uncovers two key mechanisms R&D expenditure and financial constraints through which green innovation influences firm financial outcomes [11,44]. Firms that invest more intensively in green R&D often experience a short-term decline in profitability, driven by delayed returns on innovation. In parallel, firms constrained by limited financial flexibility are less able to sustain ongoing investment in green technologies, further reducing performance [61]. These channels reveal how internal financial dynamics affect the firm’s ability to translate environmental initiatives into economic value [67,68].

The heterogeneity analysis offers additional insight into the role of environmental context. The interaction between green innovation and pollution intensity is positive and significant, indicating that firms in highly polluting sectors derive greater financial gains from green innovation. This can be attributed to stronger regulatory pressures, higher compliance costs, and increased scrutiny from stakeholders, all of which elevate the strategic value of environmental innovation [56]. In contrast, firms in low-pollution industries subject to weaker institutional pressures do not exhibit the same level of financial improvement, highlighting the contingent nature of sustainability outcomes.

Robustness checks were conducted to validate the consistency of the findings. The main results remained stable when alternative dependent variables were employed, such as ROE and total green patents. These consistent patterns suggest that the findings are not sensitive to measurement choices. To further mitigate concerns regarding endogeneity, a two-stage least squares (2SLS) estimation strategy was applied using digital transformation as an instrument. This variable, grounded in prior literature, is related to firms’ innovation behavior but not directly to short-term profitability. The 2SLS results reaffirm the baseline conclusions, and the validity of the instrument is confirmed through significant diagnostic tests.

The cumulative findings of this study contribute to a deeper and more nuanced understanding of how green innovation influences financial performance. Rather than exerting uniform effects, the impact of green innovation varies significantly across innovation intensity levels, firm capabilities, financing conditions, and regulatory environments. These insights emphasize the importance of financial preparedness, long-term innovation commitment, and institutional alignment for firms seeking to generate value from sustainability-driven strategies.

5.2. Implication

This study emphasizes the importance of government intervention to promote sustainable business practices, particularly for firms in high-emission sectors. Policymakers should provide specific incentives, such as grants, tax credits, and subsidies, to help firms in heavily polluting industries overcome the financial obstacles associated with green innovations [11,44]. These incentives can reduce short-term costs and encourage broader adoption of sustainable practices [23,28]. This study finds that strengthening environmental regulations and ensuring their uniform enforcement are essential to drive innovation and enhance corporate financial performance.

Policymakers are encouraged to enhance firms’ access to green finance, given the significant upfront costs associated with green innovation. This support can take the form of promoting financial instruments such as green bonds, green loans, and sustainability-linked credit lines, while also establishing clear green finance standards and regulatory frameworks. Facilitating access to affordable green financing is particularly important for firms in high-pollution sectors, where environmental compliance is resource-intensive. Strengthening the green finance ecosystem can help firms mitigate short-term financial burdens, support long-term innovation efforts, and better align environmental objectives with financial performance.

This research indicates that firms with higher ESG scores are better equipped to manage the economic challenges of green innovation [7,8]. Policymakers should promote comprehensive ESG reporting and integrate sustainability into business plans, offering incentives for firms that demonstrate strong ESG performance. Support for R&D activities focused on sustainable technologies is also recommended, including funding research institutions, encouraging academia-industry collaboration, and creating innovation hubs for green technology development.

The financial impacts of green innovation vary by industry, requiring sector-specific policies. High-pollution industries may need stricter regulations and targeted support for green innovation, while low-pollution industries can leverage green innovation as a market differentiator [70,71]. Policymakers should encourage long-term strategic planning by promoting frameworks that help firms assess the long-term benefits and risks of green innovation, acknowledging the nonlinear relationship between green innovation and financial performance.

By implementing these policy measures, governments can foster a supportive environment that helps firms overcome the financial challenges of green innovation. This will not only enhance the competitiveness of firms but also contribute to broader environmental and social goals, such as reducing emissions and promoting resource efficiency.

6. Conclusions

This study provides empirical evidence on the complex and dynamic relationship between green innovation and corporate financial performance in emerging market contexts, using panel data from Chinese A-share listed firms across high-pollution industries between 2009 and 2022. While green innovation is associated with short-term financial strain, particularly due to increased R&D costs and constrained liquidity, the analysis reveals a nonlinear, U-shaped relationship indicating that sustained investment leads to long-term performance improvements.

By incorporating both moderating and mediating mechanisms, this study highlights that the financial implications of green innovation are conditional on firm-specific factors such as cash flow availability, ESG performance, and innovation intensity. Moreover, the heterogeneity analysis confirms that the environmental regulatory context plays a pivotal role in amplifying the financial returns from green innovation, particularly in high-pollution industries.

These findings contribute to the literature by integrating environmental strategy with financial outcomes through a multidimensional lens, advancing the understanding of how firms can align sustainability goals with profitability. Future research may explore sector-specific policy impacts, cross-country comparisons, or the role of emerging technologies in accelerating the transition toward sustainable business models.